Speedy greedy feature selection: better redshift estimation via massive parallelism

TY - GEN

T1 - Speedy greedy feature selection

T2 - European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning 2014

AU - Gieseke, Fabian Cristian

AU - Polsterer, Kai Lars

AU - Oancea, Cosmin Eugen

AU - Igel, Christian

PY - 2014

Y1 - 2014

N2 - Nearest neighbor models are among the most basic tools in machine learning, and recent work has demonstrated their effectiveness in the field of astronomy. The performance of these models crucially depends on the underlying metric, and in particular on the selection of a meaningful subset of informative features. The feature selection is task-dependent and usually very time-consuming. In this work, we propose an efficient parallel implementation of incremental feature selection for nearest neighbor models utilizing nowadays graphics processing units. Our framework provides significant computational speed-ups over its sequential single-core competitor of up to two orders of magnitude. We demonstrate the applicability of the overall scheme on one of the most challenging tasks in astronomy: redshift estimation for distant galaxies.

AB - Nearest neighbor models are among the most basic tools in machine learning, and recent work has demonstrated their effectiveness in the field of astronomy. The performance of these models crucially depends on the underlying metric, and in particular on the selection of a meaningful subset of informative features. The feature selection is task-dependent and usually very time-consuming. In this work, we propose an efficient parallel implementation of incremental feature selection for nearest neighbor models utilizing nowadays graphics processing units. Our framework provides significant computational speed-ups over its sequential single-core competitor of up to two orders of magnitude. We demonstrate the applicability of the overall scheme on one of the most challenging tasks in astronomy: redshift estimation for distant galaxies.

M3 - Article in proceedings

SN - 978-287419095-7

SP - 87

EP - 92

BT - ESANN 2014 proceedings

A2 - Verleysen, M.

PB - i6doc.com

Y2 - 23 April 2014 through 25 April 2014

ER -